An Overflow/Underflow-Free Fixed-Point Bit-Width Optimization Method for OS-ELM Digital Circuit

M Tsukada, H Matsutani - IEICE Transactions on Fundamentals of …, 2022 - search.ieice.org
Currently there has been increasing demand for real-time training on resource-limited IoT
devices such as smart sensors, which realizes standalone online adaptation for streaming …

Virtual Screening on FPGA: Performance and Energy versus Effort

TV Aa, T Haber, TJ Ashby, R Wuyts… - arXiv preprint arXiv …, 2022 - arxiv.org
With their widespread availability, FPGA-based accelerators cards have become an
alternative to GPUs and CPUs to accelerate computing in applications with certain …

A study of the effects and benefits of custom-precision mathematical libraries for HPC codes

E Brun, D Defour, P de Oliveira Castro… - … on Emerging Topics …, 2021 - ieeexplore.ieee.org
Mathematical libraries are typically developed for use with the fixed-width data-paths on
processors and target common floating-point formats such as IEEE binary32 and binary64 …

[PDF][PDF] Virtual Screening on FPGA: Performance and Energy versus Effort

T Vander Aa, T Haber, TJ Ashby, R Wuyts… - arXiv preprint arXiv …, 2022 - academia.edu
With their wide-spread availability, FPGA-based accelerators cards have become an
alternative to GPUs and CPUs to accelerate computing in applications with certain …

Datapath Optimization for Embedded Signal Processing Architectures utilizing Design Space Exploration

J Knödtel, M Reichenbach - Proceedings of the DroneSE and RAPIDO …, 2023 - dl.acm.org
According to literature, designers spend up to 30% of the design time on optimizing data
representations in signal processing architectures [13]. Reference implementations, mostly …

[PDF][PDF] Vector Unit for Deeply-embedded, Low-power RISC-V Processors

ASS Allam - 2024 - ntnuopen.ntnu.no
Dagens smarte verden er avhengig av at små datamaskiner bygges inn i ulike produkter for
å forbedre folks liv. Disse dypt innebygde enhetene forventes i økende grad å gi høy ytelse …

Contributions to the scalability of automatic precision tuning

VP Ha - 2023 - theses.hal.science
Energy consumption is one of the major issues in computing today, shared by all domains of
computer science, from high-performance computing to embedded systems. In recent years …

FPGA acceleration of bit-true simulations for word-length optimization

J Hormigo, G Caffarena - 2021 IEEE 28th Symposium on …, 2021 - ieeexplore.ieee.org
The end of Moore's law and the arrival of new highly demanding applications have
awakened the interest in exploring different number representation formats and also …

[HTML][HTML] Research Themes

O Sentieys - people.rennes.inria.fr
Research | Olivier Sentieys Olivier Sentieys Univ. Rennes, INRIA, IRISA Skip to content Home
Activities Contacts Research Publications PhD Projects Talks Teaching TARAN group …

FPGA acceleration of bit-true simulations for word-length optimization.

J Hormigo-Aguilar, G Caffarena - 2021 - riuma.uma.es
The end of Moore's law and the arrival of new highly demanding applications have
awakened the interest in exploring different number representation formats and also …